Robotic ‘Chef’ Trained to Taste Food

A group of researchers at the University of Cambridge has created a robot “chef’ that is trained to taste food at different stages of the chewing process. While it is tasting the food, the robot determines if it’s sufficiently seasoned. 

The resulting research was published in Frontiers in Robotics & AI. 

Automating the Chewing Process

The research team worked with Beko, a domestics appliance manufacturer, to train the robot chef to assess the saltiness of a dish during the chewing process. This process is similar to what humans carry out. 

The new advancements could help develop automated or semi-automated food preparation technologies, with robots being able to determine what tastes good or doesn’t taste good. 

Humans go through a complex process when chewing food where we notice changes in texture and taste. The robotic counterpart tasted nine variations of scrambled eggs and tomatoes, and it did so during three different stages of the chewing process. The result was ‘taste maps’ of the different dishes. 

The ‘taste as you go’ approach was demonstrated to improve the robot’s ability to quickly and accurately assess the saltiness of a dish when compared to existing electronic tasting technologies. The robot had also previously been trained to make omelets based on human taster feedback. 

Taste is highly dependent on each individual’s own preferences, and the best cooks rely on their sense of taste to balance flavors while tasting as they go. 

Grzegorz Sochacki is the paper’s first author from Cambridge’s Department of Engineering. 

“Most home cooks will be familiar with the concept of tasting as you go — checking a dish throughout the cooking process to check whether the balance of flavors is right,” said Sochacki. “If robots are to be used for certain aspects of food preparation, it’s important that they are able to ‘taste’ what they’re cooking.”

Dr. Arsen Abdulali is co-author also from the Department of Engineering. 

“When we taste, the process of chewing also provides continuous feedback to our brains,” said Dr. Abdulali. “Current methods of electronic testing only take a single snapshot from a homogenized sample, so we wanted to replicate a more realistic process of chewing and tasting in a robotic system, which should result in a tastier end product.”

The team of researchers belong to Cambridge’s Bio-Inspired Robotics Laboratory run by Professor Fumiya Iida of the Department of Engineering. This lab works to train robots to carry out tasks that humans find easy but prove difficult for robots. One of these tasks is cooking. 

“We needed something cheap, small and fast to add to our robot so it could do the tasting: it needed to be cheap enough to use in a kitchen, small enough for a robot, and fast enough to use while cooking,” said Sochacki.

Constructing and Training the Robot

The researchers attached a conductance probe, which acts as a salinity sensor, to a robot arm. This enabled the robot to imitate the human process of chewing and tasting. When the researchers prepared the scrambled eggs and tomatoes dishes, they varied the number of tomatoes and amount of salt in each. 

With the probe, the robot could ‘taste’ the dishes in a grid-like fashion and process the results in seconds. 

The researchers then put the egg mixture into a blender before having the robot taste it again, which imitated the change in texture brought on by chewing. 

The technique is still a proof of concept, but the researchers believe robots will eventually be able to create food that humans will enjoy. They could even produce personalized food depending on the individual’s tastes. 

“When a robot is learning how to cook, like any other cook, it needs indications of how well it did,” said Abdulali. “We want the robots to understand the concept of taste, which will make them better cooks. In our experiment, the robot can ‘see’ the difference in the food as it’s chewed, which improves its ability to taste.”

Dr. Muhammad W. Chughtai is Senior Scientist at Beko. 

“Beko has a vision to bring robots to the home environment which are safe and easy to use,” he said. “We believe that the development of robotic chefs will play a major role in busy households and assisted living homes in the future. This result is a leap forward in robotic cooking, and by using machine and deep learning algorithms, mastication will help robot chefs adjust taste for different dishes and users.”

The research received support from Beko and the Engineering and Physical Sciences Research Council (EPSRC), which is part of UK Research and Innovation (UKRI).